Kynurenine/tryptophan ratio for detection of active tuberculosis in adults with HIV prior to antiretroviral therapy

Abstract

Objective: The aim of this study was to assess the performance of kynurenine/tryptophan ratio for tuberculosis (TB) case-finding among antiretroviral therapy (ART)-naive people with HIV (PWH), and to investigate other factors associated with kynurenine/tryptophan ratio in this population. Design: A nested case-control study based on a cohort of 812 ambulatory PWH in the Oromia region, Ethiopia. Methods: At enrolment, all participants submitted sputum samples for bacteriological TB investigations. Concentrations of kynurenine and tryptophan in plasma were quantified using liquid chromatography-mass spectrometry. Receiver operator characteristic curves were constructed to assess diagnostic performance (area under the curve; AUC) for kynurenine, tryptophan, and kynurenine/tryptophan ratio. Sensitivity, specificity, and predictive values were calculated. Kynurenine/tryptophan ratios were correlated to plasma levels of nine inflammation mediators, plasma HIV RNA levels, CD4+cell count, BMI, and mid-upper arm circumference (MUAC). Results: We included 124 individuals with HIV-TB coinfection (HIV+/TB+) and 125 with HIV mono-infection (HIV+/TB-). Tryptophan levels were lower in HIV+/TB+ than in HIV+/TB-(median 19.5 vs. 29.8 μmol/l, P < 0.01), while kynurenine levels were similar between these groups (median 2.95 vs. 2.94 μmol/l, P = 0.62). Median kynurenine/tryptophan ratio was 0.15 in HIV+/TB+, significantly higher compared with HIV+/TB-(0.11; P < 0.01), with AUC 0.70 for TB detection. Kynurenine/tryptophan ratio was positively correlated to plasma HIV RNA levels, IP-10, IL-18, and IL-27, and negatively correlated to CD4+cell count, BMI, and MUAC (all P < 0.01). Conclusion: Among ART-naive PWH, kynurenine/tryptophan ratio has modest potential for TB discrimination, limiting its utility for TB case-finding in this population.